Frequent crashes on Acer Aspire X1300

[gregrocker]

Would appreciate any read on this possible. It's my Uncle's slimline desktop that keeps crashing. He gets a White screen which hangs until he shuts it down.

Windows Update pushes through an NVidia 9200 GPU driver even though the PC has 8200 chip. I have tried rolling back to 8200 WHQL driver from 2009. There is now a newer Nvidia 331.82 Display driver recommended by Nvidia site GPU scan which I have just installed.

The NVIDIA chipset scan says it has the latest. I remedied missing co-Processor driver by browsing into the last chipset provided for this model then installed the rest of it detailed here in its ReadMe. It still seems pretty old.

This Vista 32-bit MCP78/72 nForce driver package consists of the following components:
Ethernet Driver (v67.76.1) WHQL
SataIDE Driver (v10.3.0.21) WHQL
RAID Driver (v10.3.0.21) WHQL
RAIDTOOL (v10.3.0.21)
SMBus (v4.64) WHQL
SMU (v1.39) WHQL
GPU Driver(v171.37) WHQL
Installer (v5.84)

Thanks!

 

[matts6887]

what about doing a check for disk and or system errors using sfc /scannow and or dskchk /f? Would that be a viable option(s) at this point? Cause generally speaking a crash usually means something in windows system files themselves are possibly corrupted; and or possibly in this case; the gpu is on the fritz. And im thinking that might solve the issue(S).

 

[gregrocker]

Update: After the steps I took above there was another white screen crash requiring forced shutdown. I am uploading another 7F tool report below.

Then I will be rolling back the Display driver to the 2009 Nvidia 8200 driver which was the last one for 8200 before Updates started pushing through 9200 drivers. However the one I'm rolling back is a 11/13 beta driver recommended by the Nvidia driver scan as detailed in OP.

 

[gregrocker]

Am working in Admin log resolving repeat errors and I get this crash which stays up long enough to snip it, then uninstall all Nvidia in Control Panel that is not specified a driver:

View attachment 298780
View attachment 298781

This is now what is left now of Nvidia in Control Panel. Do we need those 3D drivers which came from that latest GeForce version I already rolled back?

View attachment 298785

I'm going to uninstall all those higher version 331. drivers Nvidia scan recommended today since I already rolled them back in Dev Mgr>Display to the 17.3 Nv8200 driver from 2009.

 

[carwiz]

That probably explains the last two entries in the log:

Event[21692]:
  Log Name: Application
  Source: Desktop Window Manager
  Date: 2013-12-25T19:03:47.000
  Event ID: 9010
  Task: N/A
  Level: Information
  Opcode: N/A
  Keyword: Classic
  User: N/A
  User Name: N/A
  Computer: Paul-Acer
  Description: 
[COLOR=navy]A request to disable the Desktop Window Manager was made by process (2024)[/COLOR]
Event[21693]:
  Log Name: Application
  Source: Desktop Window Manager
  Date: 2013-12-25T19:03:47.000
  Event ID: 9013
  Task: N/A
  Level: Information
  Opcode: N/A
  Keyword: Classic
  User: N/A
  User Name: N/A
  Computer: Paul-Acer
  Description: 
[COLOR=navy]The Desktop Window Manager was unable to start because composition was disabled by a running application[/COLOR]

The MSDL server is pitifully slow tonight for some reason. (Needed for dump symbols) It hasn't responded after several tries. I may have to call it a night. One of the other BSOD guys should be showing up soon. Sorry I couldn't help.

 

[gregrocker]

Thanks Ken. I should have mentioned that TeamViewer turns off Aero to save bandwidth when I connect to this PC as I was at the time.

 

[carwiz]

Well, this certainly doesn't look like good news. The only dump in the upload indicates a hardware error. More specifically, Processor L1 Cache. This error can be caused by over heating and sometimes the processor will recover. L1 cache sits right next to the processor core and can be the first to go on over heat conditions. The system event log ended too early so I couldn't check for throttling.

*******************************************************************************
*                                                                             *
*                        Bugcheck Analysis                                    *
*                                                                             *
*******************************************************************************
[B]WHEA_UNCORRECTABLE_ERROR (124)
[/B]A fatal hardware error has occurred. Parameter 1 identifies the type of error
source that reported the error. Parameter 2 holds the address of the
WHEA_ERROR_RECORD structure that describes the error conditon.
Arguments:
Arg1: 00000000, Machine Check Exception
Arg2: 85f92024, Address of the WHEA_ERROR_RECORD structure.
Arg3: b2004000, High order 32-bits of the MCi_STATUS value.
Arg4: 64000175, Low order 32-bits of the MCi_STATUS value.
Debugging Details:
------------------

BUGCHECK_STR:  0x124_AuthenticAMD
CUSTOMER_CRASH_COUNT:  1
DEFAULT_BUCKET_ID:  VISTA_DRIVER_FAULT
PROCESS_NAME:  YahooMessenger
CURRENT_IRQL:  1f
STACK_TEXT:  
82b37d2c 82e34fcd 00000124 00000000 85f92024 nt!KeBugCheckEx+0x1e
82b37d68 82ae47ac 85f90821 85f92024 82a8a790 hal!HalBugCheckSystem+0xab
82b37d98 82e3527f 85f83438 82e42e40 00000000 nt!WheaReportHwError+0x230
82b37eac 82e35371 85f83438 85f83488 00000000 hal!HalpMcaReportError+0x3b
82b37ed0 82e355dd 00000000 85f83488 82b37ff0 hal!HalpMceHandler+0x8b
82b37ee0 82e2b5f5 00000000 00000000 00000000 hal!HalpHandleMachineCheck+0x3b
82b37ff0 82e28261 00000000 801dc000 00000000 hal!HalpMcaExceptionHandler+0x87
82b37ff0 00000000 00000000 801dc000 00000000 hal!HalpMcaExceptionHandlerWrapper+0x89

STACK_COMMAND:  kb
FOLLOWUP_NAME:  MachineOwner
MODULE_NAME: hardware
IMAGE_NAME:  hardware
DEBUG_FLR_IMAGE_TIMESTAMP:  0
FAILURE_BUCKET_ID:  0x124_AuthenticAMD_PROCESSOR_CACHE
BUCKET_ID:  0x124_AuthenticAMD_PROCESSOR_CACHE
Followup: MachineOwner
---------
0: kd> !errrec 85f92024
===============================================================================
Common Platform Error Record @ 85f92024
-------------------------------------------------------------------------------
Record Id     : 01cea1f3c3500041
Severity      : Fatal (1)
Length        : 928
Creator       : Microsoft
Notify Type   : Machine Check Exception
Timestamp     : 8/26/2013 0:48:56
Flags         : 0x00000000
===============================================================================
Section 0     : Processor Generic
-------------------------------------------------------------------------------
Descriptor    @ 85f920a4
Section       @ 85f9217c
Offset        : 344
Length        : 192
Flags         : 0x00000001 Primary
Severity      : Fatal
Proc. Type    : x86/x64
Instr. Set    : x86
Error Type    : Cache error
Operation     : Generic
Flags         : 0x00
Level         : 1
CPU Version   : 0x0000000000100f23
Processor ID  : 0x0000000000000000
===============================================================================
Section 1     : x86/x64 Processor Specific
-------------------------------------------------------------------------------
Descriptor    @ 85f920ec
Section       @ 85f9223c
Offset        : 536
Length        : 128
Flags         : 0x00000000
Severity      : Fatal
Local APIC Id : 0x0000000000000000
CPU Id        : 23 0f 10 00 00 08 02 00 - 09 20 80 00 ff fb 8b 17
                00 00 00 00 00 00 00 00 - 00 00 00 00 00 00 00 00
                00 00 00 00 00 00 00 00 - 00 00 00 00 00 00 00 00
Proc. Info 0  @ 85f9223c
===============================================================================
Section 2     : x86/x64 MCA
-------------------------------------------------------------------------------
Descriptor    @ 85f92134
Section       @ 85f922bc
Offset        : 664
Length        : 264
Flags         : 0x00000000
Severity      : Fatal
Error         : [B][COLOR=red]DCACHEL1_EVICT_ERR[/COLOR][/B] (Proc 0 Bank 0)
  Status      : 0xb200400064000175

 

[gregrocker]

I don't think the dust has ever been cleaned out in 5-6 years, and the tower sits on a counter right behind a stove where a lot of grease cooking is done. So it needs to be opened the next time I'm there in Feb. to see what kind of mess is inside. What can I have him do til then?

His roommate has the same model which is the one I focused on cleaning since he's a chain smoker and everything was coated with soot. When he cleaned the CPU fan it went out which required ordering a new one from China, but he has good performance now. But he does also get periodic crashes that I thought we'd traced before to the wrong Nvidia 9200 graphics driver being pushed through from Updates. When I'd roll it back to the correct 2009 Nvidia 8200 driver the crashes would stop. Every time I get reports of crashes I check and the 9200 driver has slipped in, even though hidden. Rolling back usually helps.

 

[carwiz]

Also - I did locate some info on Hal.dll related hardware errors.

1. It suggests checking for a BIOS update.
2. Set the BIOS back to default settings and clear CMOS memory. Clearing CMOS memory forces the BIOS to re-evaluate hardware and re-block CMOS.
3. Remove OS to BIOS software such as utilities from manufacturer's like Asus' AI Suite. Intel has one called Management Interface. Some of these may not be compatible with older BIOS versions.

 

[gregrocker]

Thank you again Ken and sorry I've been swamped here. I'll try to connect via TeamViewer to get the latest dump files from crashes which persist, then install SpeedFan to monitor temps. But the BIOS reset may need to wait til I am there in Feb when I'll also clean out the case good.

There's no management interface software on there I know of except the drivers for the Nvidia NForce 430 chipset. Other than this do you think we can eliminate software unless something shows up in the new dump?

 

[carwiz]

No problem here Greg; I'm still subscribed.
Yeah, we'll look at any new BSODs. Is the system (cabinet) cleaned out? Hardware errors are tough to pin down because they depend on the data going in and out of them. If that data is corrupt, hardware fails. And the reverse is true too. So the way to find it is to eliminate each possible cause, including software.

 

[gregrocker]

The Acer slimline has sat for three years behind his monitor directly in front of a stove where lots of grease cooking is done. It was either here or on a dusty floor under the counter. I wonder if grease has built up which would probably be a similar challenge to clean up as the roommate same model in a den where he chain-smoked (until recently) and we had to scrape the tar off CPU and fan with qtips. I think dust would have been better since tar and now grease can't be blown out, and when dust is added makes gunk.

I'll clean it as soon as I arrive back in Calif for 3 weeks Feb 4th.

Shouldn't the BIOS shut down the PC as a clear warning of overheating, before it white screens instead? Can we adjust this? I think it's the most basic BIOS.

All I find on overheating is in this review ProductWiki: Acer Aspire X1300 - Desktops :

Acer Aspire X1300

$475.00 Released January, 2009

The Pros: Affordable prices. Slick aesthetic. Powerful enough for the majority of consumers - ample processing power, storage capacity and RAM.

The Cons: The fan overheats which causes the screen to go blank. then have to resart. Tight spacing inside to upgrade components. Comes bundled with lots of crapware.
The X1300 is an entry level desktop PC that is designed for entertainment and home computing from Acer's 2009 Aspire lineup. The PC is a step up from the X1200 since it features a more powerful 2.3GHz AMD Phenom X4 Quad-Core processor, up to 4GB of DDR2 memory, and comes with an up to 640GB hard drive.

I have various size case fans there Ill try to fit in as well, but the space is almost nil.

 

[carwiz]

There's normally options in the BIOS for alarms and fan speed ranges but I don't think it shuts down the PC. Intel processors have a feature that will throttle the CPU core(s) when the temperature gets high enough but I don't think AMDs do this. AMDs are known to run until they go up in smoke.

Clean fans make a huge difference in their ability to move air. You'd be surprised how little "crud" it takes to reduce the flow by half. An additional fan, if there's room, would help but the existing fans will most likely need cleaning. Be sure to check the power supply fan too. It's part of the thermal design for most small systems.

 

[gregrocker]

I'm installing Core Temp to easily monitor temps in SysTray, and set to Hibernate if it reaches max.

View attachment 301015

Here's the latest crash file:

 

[carwiz]

Well, there's no dump in that group and the system event log ends in November of 2013. The application event log shows Desktop Manager crashing but that could be it's just in memory if that's when the BSOD occurred. You might want to go into the events logs and clear them. I always thought Windows would overlay when it ran out of space but I've seen several systems like this.

This is the Display driver info. Did you check for update on it from Acer? I didn't reload the old zip so I'm working off memory. (Dangerous.)

Adapter Type GeForce 9200, NVIDIA compatible
Adapter Description NVIDIA GeForce 9200
Adapter RAM 256.00 MB (268,435,456 bytes)
Installed Drivers nvd3dum.dll,nvwgf2um.dll,nvwgf2um.dll
Driver Version 9.18.13.3165
INF File oem16.inf (Section046 section)
Driver c:\windows\system32\drivers\nvlddmkm.sys ([B]9.18.13.3165[/B], 9.93 MB (10,410,272 bytes), [B]10/27/2013[/B] 10:13 AM)

And this is the BIOS. Is this current?

BIOS Version/Date Phoenix Technologies, [B]LTD R01-B3, 4/1/2009[/B]

 

[gregrocker]

There is no BIOS update from Acer. The Nvidia 9200 driver was discussed earlier and is not the actual chip installed - it's an 8200. But in 2009 the last 8200 driver was posted and since then Updates has been pushing through 9200 driver. It works fine without any recent crashed on roommates identical machine. We've also tried all legacy Nvidia drivers for 8200 and 9200. The chipset is Nvidia but doesn't seem to install anything when its run.

Since it's still crashing how do we get a current dump file? It said it found the Dump file when I ran the tool. If that was the old one should we delete it and wait for another crash?

 

[carwiz]

I would leave the last couple of dumps. They may be good to refer to if the next one is inconclusive. Not sure why the utility didn't pick up the dump(s). I've seen uploads with 10 or more dumps. The "utility" is a group of scripts that retrieve the files so you need to give it time to complete. I would allow at least 5 minutes to be on the safe side.

Also: Have you tried the 9200 driver, if it's different date. That number is what's reported by the chip set but you can verify that by looking at Hardware IDs.

Here's an example:

 

[gregrocker]

Update: Arrove here a week ago and immediately cleaned out greasy dust gunk from heatsink and repasted processor. Then did a bare bones reinstall only supplying the missing co-processor driver by browsing into Acer chipset - think it was Mgmt Engine.

Ran all Windows Updates but didn't import factory chipset, HP printer software or any third party drivers at all. Windows gave the Photosmart C4600 a driver that allows Windows scanning but saves all scan's as .pdf opened by Picture viewer - anyone know a way around this?

PC now runs ridiculously cool 78-98F where it was at 150F min. But as it hadn't been running all that hot I am still flummoxed as to the exact cause of the white screen crashes. But there hasn't been one since the cleaning and reinstall.

 

[carwiz]

I'm still subscribed. If you get a dump, post it.